Parkinson’s disease (PD) patients’ well-being and clinical trial success is critically reliant on understanding disease course, as the rate of deterioration varies dramatically between patients. The genetic architecture of PD progression and prognosis has not been defined and most genome-wide analyses do not capture the time dimension. Using genome-wide survival studies of deeply phenotyped, longitudinally assessed PD cohorts, we aim to uncover the genetic architecture of PD progression to dementia (PDD) over time.
In Phase 1, we examined 11.2 million variants and 31,578 longitudinal visits from 3,821 PD patients from 15 international cohorts. In Phase 2, we are currently scaling up this search to >60,000 longitudinal visits and >10,000 patients from >25 cohorts. Cox proportional hazard analysis in Phase 1 identified RIMS2 variant, rs182987047[A>T], to be significantly associated with accelerated progression to PDD (discovery: hazard ratio (HR)=4.74, P=1.16E-09; replication: HR=6.20, P=0.004), and TMEM108 and WWOX variants, rs138073281[A>C] (meta-analysis: HR= 2.86, P=2.09E-08) and rs8050111[A>G] (meta-analysis: HR=2.12, P=2.37E-08) were suggestively associated, respectively. Generalized linear mixed model assessments of Mini Mental State Exam scores confirmed that RIMS2 rs182987047 carriers [T] cognitively declined more rapidly over time compared to non-carriers [A] (P=0.0014). Moreover, we confirmed associations for GBA (HR=1.93, P=0.0002) and APOE (HR=1.48, P=0.001). Finally, polygenic hazard scores exhibited a substantial aggregate association with increased dementia risk.
This collaborative search is beginning to systematically clarify genetic drivers of PD progression at genome scale. Emerging results raise the provocative hypothesis that the determinants of Parkinson’s susceptibility and progression are importantly different.